Papers by Dongdong Zhang

Zero-Shot Cross-Lingual Transfer of Neural Machine Translation with Multilingual Pretrained Encoders (2021.emnlp-main)

Challenge:	Named entity recognition (NER) suffers from the scarcity of annotated training data, especially for low-resource languages without labeled data.
Approach:	They propose a cross-lingual entity projection framework to enable zero-shot cross-linguistic NER with the help of a multilingual labeled sequence translation model.
Outcome:	The proposed method outperforms the baseline method on two benchmarks by a large margin of +3 7 F1 scores and achieves state-of-the-art performance.

Copied to clipboard

Guanhua Chen, Shuming Ma, Yun Chen, Li Dong, Dongdong Zhang, Jia Pan, Wenping Wang, Furu Wei

Discourse-Centric Evaluation of Document-level Machine Translation with a New Densely Annotated Parallel Corpus of Novels (2023.acl-long)

Challenge:	Existing work on improving cross-lingual transferability of NMT model is under-explored.
Approach:	They propose a model that leverages a multilingual pretrained encoder to improve cross-lingual transferability.
Outcome:	The proposed model outperforms mBART and m2m-100 on a zero-shot cross-lingual transfer task.

Copied to clipboard

Yuchen Eleanor Jiang, Tianyu Liu, Shuming Ma, Dongdong Zhang, Mrinmaya Sachan, Ryan Cotterell

ShifCon: Enhancing Non-Dominant Language Capabilities with a Shift-based Multilingual Contrastive Framework (2025.acl-long)

Challenge:	Several recent papers claim to have achieved human parity at sentence-level machine translation.
Approach:	They propose to use a dataset with rich discourse annotations to evaluate MT performance . they find that MT outputs differ fundamentally from human translations in terms of latent discourse structures.
Outcome:	The proposed dataset builds upon the large-scale parallel corpus BWB . it covers 15,095 entity mentions in both languages and compares them to human translations .

Copied to clipboard

Hengyuan Zhang, Chenming Shang, Sizhe Wang, Dongdong Zhang, Yiyao Yu, Feng Yao, Renliang Sun, Yujiu Yang, Furu Wei

PAEG: Phrase-level Adversarial Example Generation for Neural Machine Translation (2022.coling-1)

Challenge:	Experiments show that ShifCon significantly enhances the performance of non-dominant languages due to the imbalance in training data across languages.
Approach:	They propose a Shift-based multilingual Contrastive framework that aligns the internal forward process of other languages toward that of the dominant one.
Outcome:	The proposed framework significantly improves performance of non-dominant languages, particularly for low-resource ones.

Copied to clipboard

Juncheng Wan, Jian Yang, Shuming Ma, Dongdong Zhang, Weinan Zhang, Yong Yu, Zhoujun Li

Respond in my Language: Mitigating Language Inconsistency in Response Generation based on Large Language Models (2024.acl-long)

Challenge:	Existing methods for adversarial example generation are word-level or character-level, which ignore the ubiquitous phrase structure.
Approach:	They propose a phrase-level adversarial example generation framework to enhance the robustness of the translation model by adopting a sentence-level substitution strategy.
Outcome:	The proposed method improves translation performance and robustness to noise on three benchmarks.

Copied to clipboard

Liang Zhang, Qin Jin, Haoyang Huang, Dongdong Zhang, Furu Wei

Multilingual Agreement for Multilingual Neural Machine Translation (2021.acl-short)

Challenge:	Large Language Models (LLMs) show strong instruction understanding ability across multiple languages, but are easily biased towards English in instruction tuning.
Approach:	They propose to use a model with Pseudo-Inconsistent Penalization to prevent the model from generating English responses when given non-English language prompts during training and prior Enhanced decoding to improve the language consistency of the model.
Outcome:	The proposed methods significantly improve the language consistency of the model without multilingual data.

Copied to clipboard

Jian Yang, Yuwei Yin, Shuming Ma, Haoyang Huang, Dongdong Zhang, Zhoujun Li, Furu Wei

Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Models (2026.findings-acl)

Challenge:	Existing models that only use auxiliary languages to encourage multilingual agreement ignore the relationships between different language pairs.
Approach:	They propose a multilingual agreement-based method which explicitly models the agreement between different translation directions by randomly substituting some fragments of the source language with their counterpart translations of auxiliary languages.
Outcome:	The proposed method improves on the multilingual translation task of 10 language pairs.

Copied to clipboard

Hengyuan Zhang, Zhihao Zhang, Ercong Nie, Mingyang Wang, Zunhai Su, Yiwei Wang, Qianli Wang, Shuzhou Yuan, Xufeng Duan, Qibo Xue, Zeping Yu, Chenming Shang, Xiao Liang, Jing Xiong, Hui Shen, Chaofan Tao, Zhengwu Liu, Senjie Jin, Zhiheng Xi, Dongdong Zhang, Sophia Ananiadou, Tao Gui, Ruobing Xie, Hayden Kwok-Hay So, Hinrich Schuetze, Xuanjing Huang, Qi Zhang, Ngai Wong

Explainability & Interpretability in NLP

Challenge:	Existing literature on mechanistic interpretation (MI) treats it as an observational science, leaving practical applications underexplored.
Approach:	They propose a survey structured around the pipeline to identify and improve MI models.
Outcome:	The proposed framework enables tangible improvements in Alignment, Capability, and Efficiency.

Revamping Multilingual Agreement Bidirectionally via Switched Back-translation for Multilingual Neural Machine Translation (2024.findings-eacl)

Copied to clipboard

Hongyuan Lu, Haoyang Huang, Dongdong Zhang, Furu Wei, Wai Lam

Not All Languages Are Created Equal in LLMs: Improving Multilingual Capability by Cross-Lingual-Thought Prompting (2023.findings-emnlp)

Challenge:	Current multilingual agreement (MA) methods require parallel data between multiple language pairs, which is not always realistic and optimize the agreement in an ambiguous direction, which hampers the translation performance.
Approach:	They propose a novel multilingual agreement framework that optimizes agreement bidirectionally with the Kullback-Leibler Divergence loss.
Outcome:	The proposed method improves strong baselines on the task of multilingual neural machine translation with three benchmarks: TED Talks, News, and Europarl.

Copied to clipboard

Haoyang Huang, Tianyi Tang, Dongdong Zhang, Xin Zhao, Ting Song, Yan Xia, Furu Wei

Low-Resource NLP

Challenge:	Large language models (LLMs) demonstrate impressive multilingual capability, but their performance varies substantially across different languages.
Approach:	They propose a generic template prompt that stimulates cross-lingual and logical reasoning skills to enhance task performance across languages.
Outcome:	The proposed method improves multilingual capability across languages and covers high-resource and low-resourced languages.

From Word to World: Can Large Language Models be Implicit Text-based World Models? (2026.acl-long)

Copied to clipboard

Yixia Li, Hongru Wang, Jiahao Qiu, Zhenfei Yin, Dongdong Zhang, Cheng Qian, Zeping Li, Xiaoteng Ma, Guanhua Chen, Heng Ji

Smart-Start Decoding for Neural Machine Translation (2021.naacl-main)

Challenge:	Agentic learning increasingly hinges on interaction, yet real-world experience is expensive, limited, and often irreversible at inference time.
Approach:	They propose a framework that reframes language modeling as next-state prediction under interaction.
Outcome:	The proposed framework evaluates world models in text-based environments . it shows that sufficiently trained models capture coherent environment dynamics .

Copied to clipboard

Jian Yang, Shuming Ma, Dongdong Zhang, Juncheng Wan, Zhoujun Li, Ming Zhou

GanLM: Encoder-Decoder Pre-training with an Auxiliary Discriminator (2023.acl-long)

Challenge:	Existing neural machine translation models adopt a monotonic decoding order of either left-to-right or right-to left.
Approach:	They propose a method that starts decoding target words from the right side of a median word and generates words on the left.
Outcome:	The proposed method outperforms baseline models on three datasets.

Copied to clipboard

Jian Yang, Shuming Ma, Li Dong, Shaohan Huang, Haoyang Huang, Yuwei Yin, Dongdong Zhang, Liqun Yang, Furu Wei, Zhoujun Li

A Simple and Effective Unified Encoder for Document-Level Machine Translation (2020.acl-main)

Challenge:	Existing pre-training methods underutilize the benefits of language understanding for generation.
Approach:	They propose a GAN-style model for encoder-decoder pre-training with an auxiliary discriminator.
Outcome:	The proposed model outperforms existing pre-trained models and achieves state-of-the-art performance.

Copied to clipboard

Shuming Ma, Dongdong Zhang, Ming Zhou

TRIP: Accelerating Document-level Multilingual Pre-training via Triangular Document-level Pre-training on Parallel Data Triplets (2023.findings-emnlp)

Challenge:	Existing models for document-level machine translation use two separate encoders to model the source sentences and document- level contexts.
Approach:	They propose a unified encoder that can outperform existing models of dual-encoder models . they propose to use document-level contexts to model the interaction between the contexts and the source sentences .
Outcome:	The proposed model outperforms baseline models of dual-encoder models in terms of BLEU and METEOR scores.

Copied to clipboard

Hongyuan Lu, Haoyang Huang, Shuming Ma, Dongdong Zhang, Wai Lam, Zhaochuan Gao, Anthony Aue, Arul Menezes, Furu Wei

Zero-shot Cross-lingual Transfer of Prompt-based Tuning with a Unified Multilingual Prompt (2022.emnlp-main)

Challenge:	Existing approaches to multilingual sequence-to-sequence pre-training rely on monolingual corpora and sometimes synthetic document-level bilingual corporata.
Approach:	They propose to leverage document-level trilingual parallel corpora to improve sequence-to-sequence multilingual pre-training by using a novel method called Grafting.
Outcome:	The proposed method achieves strong state-of-the-art (SOTA) scores on three multilingual document-level machine translation benchmarks and one cross-lingual abstractive summarization benchmark.

Copied to clipboard

Lianzhe Huang, Shuming Ma, Dongdong Zhang, Furu Wei, Houfeng Wang

VFA: Empowering Multilingual MLLMs via Vision-Free Adaptation (2026.acl-long)

Challenge:	Existing work focuses on monolingual prompts, but we study multilingual prompt for multilingual models.
Approach:	They propose a model that uses a unified prompt for all languages, called UniPrompt, to alleviate the effort of designing different prompts for multiple languages.
Outcome:	The proposed model outperforms baseline models in the zero-shot cross-lingual setting.

Copied to clipboard

Yixia Li, Yaqing Shi, Zhiwen Ruan, Dongdong Zhang, Lingjie Jiang, Shaohan Huang, Yun Chen, Guanhua Chen, Furu Wei

Multimodality

Challenge:	Multimodal large language models have advanced rapidly, yet most remain English-centric . scaling multilingual multimodal instruction tuning is limited by the scarcity and high cost of non-English image–text supervision.
Approach:	They propose a framework that decouples multilingual language enhancement from visual alignment by composing complementary task vectors over a shared LLM backbone.
Outcome:	The proposed framework achieves competitive performance with a fully multimodally trained model using less than 2% of the text data.

Igniting Creative Writing in Small Language Models: LLM-as-a-Judge versus Multi-Agent Refined Rewards (2025.emnlp-main)

Copied to clipboard

Xiaolong Wei, Bo Lu, Xingyu Zhang, Zhejun Zhao, Dongdong Shen, Long Xia, Dawei Yin

Improving Neural Machine Translation with Soft Template Prediction (2020.acl-main)

Challenge:	Existing methods for enhancing Large Language Models (LLMs) struggle with novelty and Reinforcement Learning from human feedback (RLHF) is costly.
Approach:	They propose to use a Reward Model (RM) and a principle-guided LLM-as-a-Judge to enhance creative output over baselines.
Outcome:	The proposed approach significantly enhances creative output over baselines, but the principle-guided LLM-as-a-Judge yields superior generation quality.

Copied to clipboard

Jian Yang, Shuming Ma, Dongdong Zhang, Zhoujun Li, Ming Zhou

DMN: A Compositional Framework for Jailbreaking Multimodal LLMs with Multi-Image Inputs (2026.acl-long)

Challenge:	Recent advances in neural machine translation (NMT) depend on source text to generate translation.
Approach:	They propose to use extracted templates from tree structures as soft target templates to guide the translation procedure.
Outcome:	The proposed model outperforms baseline models on four benchmarks and demonstrates the effectiveness of soft target templates.

Copied to clipboard

Wenzhuo Xu, Zhipeng Wei, Zonghao Ying, Deyue Zhang, Dongdong Yang, Xiangzheng Zhang, Quanchen Zou

Responsible & Trustworthy NLP

Challenge:	Existing jailbreak methods only use a single image, restricting the attack space . Existing frameworks only use single image to distribute harmful requests across multiple images .
Approach:	They propose a compositional jailbreak framework that leverages Distributed instruction, Multimodal evidence and a Number chain task to fully enhance the jailbreak performance.
Outcome:	The proposed framework achieves attack success rates of over 90% on GPT-4o, Gemini-2.5-pro and Claude Sonnet 4 .

LVP-M3: Language-aware Visual Prompt for Multilingual Multimodal Machine Translation (2022.emnlp-main)

Copied to clipboard

Hongcheng Guo, Jiaheng Liu, Haoyang Huang, Jian Yang, Zhoujun Li, Dongdong Zhang, Zheng Cui

Document Classification for COVID-19 Literature (2020.findings-emnlp)

Challenge:	Recent advances struggle to train a separate model for each language pair, which is costly and unaffordable when the number of languages increases in the real world.
Approach:	They propose to train different MMT models to support translations between different languages.
Outcome:	The proposed model is able to handle the above issues by providing a shared semantic space for multiple languages.

Copied to clipboard

Bernal Jimenez Gutierrez, Jucheng Zeng, Dongdong Zhang, Ping Zhang, Yu Su

Information Retrieval

Challenge:	a global pandemic has made it more important than ever to quickly and accurately retrieve relevant scientific literature for effective consumption by researchers in a wide variety of fields.
Approach:	They analyze a LitCovid dataset to find out how classification models can help organize COVID-19 research papers.
Outcome:	The proposed model outperforms all baseline models on the LitCovid dataset . it also outperformed BioBERT and other models with micro-F1 and accuracy scores of 86% and 75% .

Towards Making the Most of Cross-Lingual Transfer for Zero-Shot Neural Machine Translation (2022.acl-long)

Copied to clipboard

Guanhua Chen, Shuming Ma, Yun Chen, Dongdong Zhang, Jia Pan, Wenping Wang, Furu Wei

On the Off-Target Problem of Zero-Shot Multilingual Neural Machine Translation (2023.findings-acl)

Challenge:	Existing unsupervised neural machine translation systems can degrade when labeled data is limited.
Approach:	They propose a multilingual pretraining and multilingual fine-tuning for facilitating cross-lingual transfer in zero-shot translation using a parallel dataset.
Outcome:	The proposed model outperforms state-of-the-art models on many-to-English translation by over 7.2 and 5.0 BLEU.

Copied to clipboard

Liang Chen, Shuming Ma, Dongdong Zhang, Furu Wei, Baobao Chang

BlonDe: An Automatic Evaluation Metric for Document-level Machine Translation (2022.naacl-main)

Challenge:	Despite its success, multilingual neural machine translation suffers from the off-target issue, where the translation is in the wrong language.
Approach:	They propose a language-aware vocabulary sharing algorithm that can be used to increase the lexical distance between languages by isolating the vocab of different languages in the decoder.
Outcome:	The proposed algorithm reduces off-target rate for 90 translation tasks from 29% to 8%, while improving overall BLEU score by an average of 1.9 points without extra training cost or sacrificing the supervised directions’ performance.

Copied to clipboard

Yuchen Jiang, Tianyu Liu, Shuming Ma, Dongdong Zhang, Jian Yang, Haoyang Huang, Rico Sennrich, Ryan Cotterell, Mrinmaya Sachan, Ming Zhou

Language-Specific Neurons: The Key to Multilingual Capabilities in Large Language Models (2024.acl-long)

Challenge:	Standard evaluation metrics, e.g., BLEU, TER and METEOR, focus on the quality of translations at the sentence level and do not consider discourse-level features.
Approach:	They propose to use a metric to take discourse coherence into consideration by categorizing discourse-related spans and calculating the similarity-based F1 measure of categorized spans.
Outcome:	The proposed metric possesses better selectivity and interpretability at the document-level, and is more sensitive to document- level nuances.

Copied to clipboard

Tianyi Tang, Wenyang Luo, Haoyang Huang, Dongdong Zhang, Xiaolei Wang, Xin Zhao, Furu Wei, Ji-Rong Wen

Semantic Text Processing

Challenge:	Despite the impressive multilingual capabilities demonstrated by LLMs, the understanding of how these abilities develop and function remains nascent.
Approach:	They propose a novel detection method to pinpoint language-specific neurons within LLMs by selectively activating or deactivating these neurons.
Outcome:	The proposed method can “steer” the output language of LLMs by selectively activating or deactivating language-specific neurons.

Chain-of-Dictionary Prompting Elicits Translation in Large Language Models (2024.emnlp-main)

Copied to clipboard

Hongyuan Lu, Haoran Yang, Haoyang Huang, Dongdong Zhang, Wai Lam, Furu Wei

Not All Metrics Are Guilty: Improving NLG Evaluation by Diversifying References (2024.naacl-long)

Challenge:	Large language models (LLMs) have shown surprisingly good performance in multilingual neural machine translation . yet, they struggle with translating low-resource languages.
Approach:	They propose a framework that chained multilingual dictionaries to elicit translation abilities for LLMs . they show that CoD can significantly improve LLM translation by evoking more information .
Outcome:	The proposed framework improves on ChatGPT and InstructGPT's translation abilities.

Copied to clipboard

Tianyi Tang, Hongyuan Lu, Yuchen Jiang, Haoyang Huang, Dongdong Zhang, Xin Zhao, Tom Kocmi, Furu Wei

Chain-of-Reasoning: Towards Unified Mathematical Reasoning in Large Language Models via a Multi-Paradigm Perspective (2025.acl-long)

Challenge:	Existing evaluation benchmarks with limited references may not accurately reflect the quality of the model’s hypotheses.
Approach:	They propose a method to enrich evaluation benchmarks by diversifying the expression of a single reference into multiple high-quality ones to cover the semantic space of the reference sentence as much as possible.
Outcome:	The proposed method can enhance evaluation benchmarks by diversifying the expression of reference into multiple high-quality ones to cover the semantic space of the reference sentence as much as possible.

Copied to clipboard

Yiyao Yu, Yuxiang Zhang, Dongdong Zhang, Xiao Liang, Hengyuan Zhang, Xingxing Zhang, Mahmoud Khademi, Hany Hassan Awadalla, Junjie Wang, Yujiu Yang, Furu Wei

Reasoning

Challenge:	Existing work shows that LLMs rely on single-paradigm reasoning that limits their effectiveness across diverse tasks.
Approach:	They propose a new framework that integrates multiple reasoning paradigms to enable synergistic collaboration.
Outcome:	The proposed model outperforms current SOTA models in theorem proving tasks and the MATH benchmark in arithmetic tasks.

How Does Distilled Data Complexity Impact the Quality and Confidence of Non-Autoregressive Machine Translation? (2021.findings-acl)

Copied to clipboard

Weijia Xu, Shuming Ma, Dongdong Zhang, Marine Carpuat

Improving Multilingual Neural Machine Translation with Auxiliary Source Languages (2021.findings-emnlp)

Challenge:	Prior work suggests that distilled training data is less complex than manual translations.
Approach:	They propose to use sequence-level knowledge distillation to match autoregressive models' translation quality.
Outcome:	The proposed model can match translation quality of autoregressive models with distilled training data.

Copied to clipboard

Weijia Xu, Yuwei Yin, Shuming Ma, Dongdong Zhang, Haoyang Huang